SUMMARY Rationale. The proposed project seeks methods to prevent and elucidate biomarkers of neurocognition and motor deficits associated with chronic dietary reliance on cyanogenic cassava, a staple food crop for more than 600 millions of people living in the tropics. Aim 1 will implement a novel cassava processing method (wetting method, WTM) that safely removes cyanogenic compounds from cassava flour prior to human consumption in a stratified village-cluster randomized non-inferiority trial so as to compare the effectiveness of a peer-led intervention (women training other women in the WTM) with that by community-health worker specialists (2 intervention training arms). We hypothesize that the peer-led WTM intervention is non-inferior to the specialist-led intervention at the end of the 2-year intervention period and at 1-year follow-up. Task shifting to peer-led training for WTM should it prove to be effective, would provide for a more cost-effective and sustainable means to bring this intervention to scale at a community-wide level. Primary outcomes will be the cassava cyanogenic content produced by the trainee mother (N = 200 in each intervention arm) and urinary concentrations of thiocyanate (U-SCN) in her konzo and non-konzo sibling children 5 to 12 years of age. Secondary outcomes will be the neuropsychological performance scores of children assessed using the Kaufman Assessment Battery for Children (KABC-II) for cognition and the Bruininks/Oseretsky Test (BOT-2) for motor proficiency. Aim 2 will determine whether post-intervention reductions in cassava cyanogenic content and child U-SCN are associated with changes in biomarkers of cassava neurotoxicity particularly 8,12-iso- iPF2?-VI isoprostane (oxidant marker), carbamoylated albumin fragments KVPQVSTPTLVEVSR (residues 438-452) and LDELRDEGKASSAK (residues 206-219), or homocitrulline (carbamoylating markers), and scores at the KABC-II and BOT-2 testing. We hypothesize that 1) lower cassava cyanogenic content and U- SCN throughout the two-year intervention period and at one-year follow-up, will be associated with lower oxidant and carbamoylating markers but higher KABC-II and BOT-2 testing scores; and a similar pattern will be sustained at 1-year follow-up; and 2) higher levels of oxidation and carbamoylation will be associated with poorer neuropsychological performance. Biomarkers will be monitored in serum (invasive), dried blood spots (DBS) (less invasive), or urine (non-invasive) using omic methodologies with the long-term goal of developing field-based rapid tests to monitor risk for cassava associated neurodevelopmental deficits. Aim 3 will enhance the research manpower of the konzo-affected Democratic Republic of Congo.